The Doka training run got stuck here.

[SpiderJack0516]

Search before asking

• [X] I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Specific logs：

Additional

No response

[UltralyticsAssistant]

👋 Hello @SpiderJack0516, thank you for reaching out and using YOLOv5 🚀! This is an automated response to help guide you through solving your issue, and an Ultralytics engineer will be with you shortly.

If this is a 🐛 Bug Report, please share a minimum reproducible example so we can debug it efficiently. This will help us identify any potential issues in your environment or setup.

For optimal assistance with your question, ensure that you've followed our Tips for Best Training Results. Additionally, provide any relevant details such as logs or dataset examples to support your inquiry.

Requirements

Make sure that your environment meets the following conditions:

Python>=3.8.0 with necessary packages from requirements.txt installed, and ensure you have PyTorch>=1.8.

To set up YOLOv5:

git clone https://github.com/ultralytics/yolov5  # clone the repo
cd yolov5
pip install -r requirements.txt  # install dependencies

Environments

YOLOv5 supports various environments for seamless operation:

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. Check out the GCP Quickstart Guide
• Amazon Deep Learning AMI. Refer to the AWS Quickstart Guide
• Docker Image. Explore the Docker Quickstart Guide

Status

Check the CI status for the latest pass results: .

If green, all YOLOv5 GitHub Actions are passing, confirming smooth operation of our training, validation, inference, export, and benchmarks across all platforms.

Introducing YOLOv8 🚀

Don't miss out on our latest state-of-the-art model, YOLOv8 🚀. Optimized for speed, accuracy, and user-friendliness, it's perfect for your object detection, segmentation, and classification projects.

Explore the YOLOv8 Docs to get started:

pip install ultralytics

Feel free to reach out with more details if needed! 😊

[SpiderJack0516]

To add to this, the first few days were still fine for normal training, now it's impossible to train with DDP Training.

[pderrenger]

@SpiderJack0516 please ensure your environment is up-to-date with the latest YOLOv5 version and dependencies. If the issue persists, try re-cloning the repository and reinstalling the requirements. If you need further assistance, feel free to provide more details.

[SpiderJack0516]

@pderrenger I've reset to a new environment and it's still stuck there. I suspect it's because my server is not networked.

[SpiderJack0516]

@pderrenger Now, I git pull the latest code and the question is gone.

[pderrenger]

Great to hear that updating the code resolved the issue! If you have any more questions, feel free to ask.